The command line options for training can be listed by `python train.py -h`.
### Train in local mode:
### Local Train:
```bash
python train.py \
--train_data_path data/train.txt \
2>&1 | tee train.log
```
After training pass 1 batch 40000, the testing AUC is `0.807178` and the testing
After training pass 1 batch 40000, the testing AUC is `0.801178` and the testing
cost is `0.445196`.
### Run a 2 pserver 2 trainer distribute training on a single machine
### Distributed Train
Run a 2 pserver 2 trainer distribute training on a single machine
```bash
# start pserver0
python train.py \
...
...
@@ -114,3 +115,10 @@ python infer.py \
--model_path models/ \
--data_path data/valid.txt
```
Note: The AUC value in the last log info is the total AUC for all test dataset.
## Train on Baidu Cloud
1. Please prepare some CPU machines on Baidu Cloud following the steps in [train_on_baidu_cloud](https://github.com/PaddlePaddle/FluidDoc/blob/develop/doc/fluid/user_guides/howto/training/train_on_baidu_cloud_cn.rst)
1. Prepare dataset using preprocess.py.
1. Split the train.txt to trainer_num parts and put them on the machines.
1. Run training with the cluster train using the command in `Distributed Train` above.